zELDA: fitting Lyman alpha line profiles using deep learning

We present zELDA(redshift Estimator for Line profiles of Distant Lyman-Alpha emitters), an open source code to fit (Lya) line profiles. The main motivation is provide the community with easy use and fast tool analyze Lya uniformly improve understating emitting galaxies. zELDA based on commonly used 'shell-model' pre-computed full Monte Carlo radiative transfer LyaRT. Via interpolation between these spectra addition noise, we assemble a suite realistic which train deep neural network. show that network can predict model parameters high accuracy (e.g.,.0.34 dex HI column density R=12000) thus allows significant speedup over existing fitting methods. As proof concept, demonstrate potential by 97 observed from LASD data base. Comparing fitted value measured systemic redshift sources, find determines their rest frame wavelength remarkable good 0.3A (75 km/s). predicted outflow properties luminosity equivalent width, several possible trends. For example, anticorrelation neutral hydrogen density, might be explained process within galaxies